Analysis of Factors Associated With Sagittal Balance in Normal Asymptomatic Individuals

Study Design. A retrospective study. Objective. To determine primary contributors to sagittal balance and establish a predictive equation of sagittal balance in normal asymptomatic subjects. Summary of Background Data. Sagittal balance has been verified to be associated with health-related quality of life. Although many studies have been performed to explore factors contributing to sagittal balance in various disease states, no study has been conducted in normal asymptomatic subjects in East China. Methods. Medical records of 340 asymptomatic healthy volunteers were reviewed from January 2014 to August 2015, including 311 for exploring the risk factors and predictive equation, and 29 for validation. Demographic and radiological parameters were evaluated. Correlation analysis between spinopelvic parameters was pursued. Regression analyses were performed to establish predictive radiographic parameters for sagittal balance. Paired t test was conducted to test the regression equation. Results. There was a very good correlation between T1 sagittal angle and maxTK, maxLL and SS, SS and PI, and PT and PI; a moderate correlation between maxTK and maxLL, and maxLL and PI; a weak correlation between T1 sagittal angle and age, age and maxTK, age and SS, age and PT, age and SVA, T1 sagittal angel and SVA, SVA and PT, and SVA and PI; and very weak correlation between T1 sagittal angle and maxLL, maxLL and PT, SVA and LL, and PT and SS. Age, T1 sagittal angle, maxLL, PT, and PI were primary contributors to sagittal balance, which could be predicted by the regression equation: SVA = 0.294 x age + 0.367 x T1 sagittal angle - 1.149 x maxLL-0.704 x PT + 1.378 x PI - 34.164. No significant difference was found between actual SVA and predicted SVA using our equation (P = 0.307). Conclusion. Age, T1 sagittal angle, maxLL, PT, and PI were primary contributors to maintaining sagittal balance in normal asymptomatic subjects, and sagittal balance could be predicted by calculation using the equation derived in this study.